EHR Machine Learning CDS Simulator

MIMIC-IV Trained Risk Model Real-Time Inference Loop
Clinical Presets:
Active Patient Chart PT-84920
Patient Age 68 yrs
Heart Rate (HR) 118 bpm
Systolic Blood Pressure 88 mmHg
Serum Lactate 3.8 mmol/L
White Blood Cells (WBC) 16.5 k/uL
Documented Comorbidities
Current Medications
Early Sepsis ML Risk Score
92.4 / 100
Drug Conflict Severity
High
ML Ensemble Confidence
96.2%
Suggested Clinical Action & Decision Support
Immediate ICU Sepsis Protocol & Discontinue Ibuprofen
Hypotension (88 mmHg) paired with high serum lactate (3.8 mmol/L) and elevated WBC (16.5) signals imminent septic shock. Warfarin combined with NSAID (Ibuprofen) presents an unacceptably high GI haemorrhage risk exacerbated by Amiodarone CYP2C9 inhibition.
CDS Evolution: Multi-Variable ML vs Clumsy Legacy Alarms
Legacy Single-Threshold EHR Rules
• Isolated Alert: Tachycardia > 100 bpm (doctor fatigue click-away)
• Static check: WBC elevated (no temporal trend cross-referencing)
• High false-alarm rate: 84% ignored by physicians in routine workflow
Predictive Machine Learning Engine
• Correlates Lactate + BP Drop + Comorbidity before shock manifests
• CYP450 metabolic pathway simulation for bleeding risk
• Context-aware: suppressed alarms for expected post-operative baselines
6-Hour Risk Progression & Multi-Factor Forecast Hourly ML projection
Enjoy this tool? Build your own with Super